Metastasis is the most significant contributor to mortality in breast cancer patients. In breast cancer patients, bone marrow (BM) is thought to serve as a reservoir for disseminated tumor cells (DTCs), which are the hypothesized intermediaries in the metastatic process. Our goal is to molecularly classify DTCs for predictive biomarker identification and therapeutic targeting. We have developed a strategy for identifying genes associated with DTCs and a multiple assay, Nanostring nCounter for their analysis. We hypothesize that introducing a validated set of molecular markers for the detection and molecular classification of DTCs into clinical practice for the implementation of targeted therapies will result in a halting of metastatic disease development by interrupting the metastatic cascade. Our goals are to refine and validate a 46-gene expression profile for the detection and classification of DTCs using BM specimens from a 120 patient cohort of clinical stage II/III breast cancer patients and to conduct a phase II randomized trial evaluating the effect of trastuzumab on the elimination of ERRB2-positive disseminated tumor cells and DFS in early stage Her2-negative primary breast cancer patients with ERBB2 expressing BM DTCs. Our proposal will provide proof of principle that we can identify subsets of patients who will respond to specific targeted therapies based on the predictive biomarker expression profile of their DTCs and that eliminating DTCs results in a DFS benefit. This project will result in paradigm shift for treating early stage breast cancer patients based on the DTC predictive biomarker expression profile.